The study on the detection of Amyotrophic Lateral Sclerosis (ALS) through the voice carried out by CIMNE researcher Alberto Tena has captured the interest of the media and social networks.
The research is based on the analysis of the voice with artificial intelligence models that could allow the early detection of bulbar involvement in ALS. This is a collaboration between CIMNE and the University of Lleida in which the Bellvitge Hospital participated and has consisted in the development of automated voice markers to identify the first effects of the disease.
During six months, the voices of patients from the Hospital's Functional Unit for Motor Neuron Disease were recorded to develop models that allow early detection of swallowing and speech problems.
About ALS
Amyotrophic Lateral Sclerosis (ALS) is a neurodegenerative disease that is designated as spinal when the first symptoms appear in the spine, arms or legs, or as bulbar when the deterioration of neurons begins in the medulla oblongata, located at the base of the brain. Symptoms of bulbar involvement are speech and swallowing problems. The bulbar onset of ALS is minor compared to the spinal one, but patients have a worse prognosis. Despite the differentiation in the manifestation of the first symptoms of the disease, 80% of ALS patients end up experiencing articulation problems in speech. Speech impairment can begin up to three years before ALS diagnosis, making early detection of bulbar involvement essential.
Encouraging results
With these premises, the CIMNE researcher Alberto Tena began to develop this project, which starts from the recording and acoustic analysis of the pronunciation of the five vowels of the alphabet by 45 ALS patients from the Bellvitge Hospital and 18 control people. With the acoustic analysis of the pronunciation of the vowels, some machine learning systems (artificial intelligence from supervised learning) have allowed the development of models or voice markers with very satisfactory results in the identification and differentiation of the study participants with bulbar involvement, those who did not have it and the control group. “We still have a lot of work ahead of us, but the first results we have obtained show that bulbar involvement can be detected with automatic models before it is perceptible to the human ear, and that objective measures can be established to facilitate an early and accurate diagnosis”, highlights Alberto Tena. Today up to 10% of ALS patients are not correctly diagnosed of their bulbar involvement at first.
In the media
📰 El @elpuntavui entrevista Alberto Tena, investigador del @cimne: “La intel·ligència artificial serà clau per als metges”
🔗 https://t.co/woToNwVuCZ#ELA #ArtificialIntelligence #IA #MachineLearning #AmyotrophicLateralSclerosis pic.twitter.com/SjAWVQEuBX— CRAI CampusBellvitge (@BibBellvitge) April 28, 2021
Una investigació de la Universitat de Lleida permet fer un diagnòstic precoç de l’ELA a través de la veu | Diari de la Discapacitat https://t.co/PlEEhvTTQ4 @UdL_info @cimne @hbellvitge @fmiquelvalls @cemcat_em @FundacioEM @universitatscat #ELA #recerca #veu #diagnostic pic.twitter.com/wmKVWGBQ1F
— diari de la DisCapacitat (@DiariDiscapCat) April 18, 2021
📰 Un estudi de la @UdL_info i del @cimne dissenya un sistema automatitzat basat en #IA i #MachineLearning per identificar precoçment, a través de la veu, persones afectades d'#ELA amb una precisió del 95,8%.https://t.co/hpxlZBE8XJ
— Biocat (@biocat_cat) April 16, 2021
🔴 Una recerca @UdL_info
demostra que amb la veu (dient les vocals) es pot detectar l'ELA.👉 Ara es vol esbrinar si també amb la veu es pot detectar la #Covid.
Amb Francesc Solsona @EPS_UdL i Alberto Tena @cimne 👇📻 https://t.co/7BCjhFuyB7 via @SERLleida pic.twitter.com/bzDD88KJqH
— SER Ràdio Lleida (@SERLleida) April 15, 2021
🆕 Investigadors de la @UdL_info han desenvolupat un model automatitzat per identificar, a través de la #veu, les afectacions primerenques de l'esclerosi lateral amiotròfica (#ELA) en el bulb raquidi.
👉 https://t.co/WzAhEBpsEW #deteccióprecoç #IA pic.twitter.com/HmkWa1P9fF
— Recerca i Transferència UdL (@UdL_RDI) April 9, 2021
